## Define model

stancode_alpha_delta <- "

data {

int<lower = 1> N_obs; // Number of observations
int<lower = 1> N; // Number of mps
int<lower = 1> D; // Number of debates
int<lower = 1> K; // Number of topics
int<lower = 1> P; // Number of parties
int nId[N_obs];  // Unique mp id
int dId[N_obs]; // Unique debate id
int Y[N_obs]; // Response
row_vector[K] X[N_obs]; // Topic proportions
real X_sig[N_obs]; // Signatures
real Xi_margin[N]; // Electoral margin of MP i
real Xi_new_mp[N]; // Is MP i new to the House?
real Xi_cabinet[N]; // Has MP i served in the cabinet during this parliamentary term?
matrix[N,P] Xi_party_matrix; // Has MP i served in the cabinet during this parliamentary term?
real Xd_petition[D]; // Is this a petition debate?

}

parameters {

real alpha_0; // Grand intercept
real alpha[N]; // MP fixed-effect
real delta[D]; // Debate fixed-effect
real gamma[N]; // Signature coefficients (one per MP)

real<lower = 0> sigma_alpha; // alpha variance
real<lower = 0> sigma_delta; // delta variance
real<lower = 0> sigma_gamma; // gamma variance

real mu_gamma; // second level signature coefficent intercept
real mu_gamma_margin; // second level slope for electoral margin
real mu_gamma_new_mp; // second level slope for new mp
real mu_gamma_cabinet; // second level slope for minister
vector[P] mu_gamma_party; 

real mu_alpha; // second level mp random effect intercept
real mu_alpha_new_mp; // second level intercept for new mp
real mu_alpha_cabinet; // second level intercept for cabinet minister
real mu_alpha_margin; // second level intercept for electoral margin
vector[P] mu_alpha_party; 

real mu_delta; // second level debate random effect intercept
real mu_delta_petition; // second level debate effect for petition debates


}

model {
real mu[N_obs];

for(i in 1:N_obs)  {

mu[i] = (alpha_0 + 
alpha[nId[i]] + 
delta[dId[i]] + 
gamma[nId[i]] * X_sig[i]);

}



Y ~ bernoulli_logit(mu);

// Prior for grand intercepts
alpha_0 ~ normal(0, 10);

// Priors for debate intercepts
for(d in 1:D){
delta[d] ~ normal(mu_delta + mu_delta_petition*Xd_petition[d], sigma_delta);
}


// Prior for MP-effects and signature coefficients
for(m in 1:N){
alpha[m] ~ normal(mu_alpha + mu_alpha_new_mp*Xi_new_mp[m] + mu_alpha_cabinet*Xi_cabinet[m] + mu_alpha_margin*Xi_margin[m] + dot_product(mu_alpha_party, Xi_party_matrix[m,]), sigma_alpha);
gamma[m] ~ normal(mu_gamma + mu_gamma_margin*Xi_margin[m] + mu_gamma_new_mp*Xi_new_mp[m] + mu_gamma_cabinet*Xi_cabinet[m] + dot_product(mu_gamma_party, Xi_party_matrix[m,]), sigma_gamma);
}

// Hyperpriors
mu_alpha ~ normal(0, 2);
mu_alpha_new_mp ~ normal(0, 2);
mu_alpha_cabinet ~ normal(0, 2);
mu_alpha_margin ~ normal(0, 2);
for(p in 1:P) mu_alpha_party[p] ~ normal(0, 2);

mu_gamma ~ normal(0, 2);
mu_gamma_margin ~ normal(0, 2);
mu_gamma_new_mp ~ normal(0, 2);
mu_gamma_cabinet ~ normal(0, 2);
for(p in 1:P) mu_gamma_party[p] ~ normal(0, 2);

mu_delta ~ normal(0, 2);
mu_delta_petition ~ normal(0, 2);

}


generated quantities{

real log_likelihood;
real deviance;
log_likelihood = 0;
for(i in 1:N_obs)  {

log_likelihood = log_likelihood + bernoulli_logit_lpmf(Y[i] | (alpha_0 + 
alpha[nId[i]] +
delta[dId[i]] +
gamma[nId[i]] * X_sig[i]));

}

deviance = -2*log_likelihood;

}

"
